Filters








5,018 Hits in 2.8 sec

Privacy-Preserving Multiple Linear Regression of Vertically Partitioned Real Medical Datasets

Hiroaki Kikuchi, Chika Hamanaga, Hideo Yasunaga, Hiroki Matsui, Hideki Hashimoto, Chun-I Fan
2018 Journal of Information Processing  
Our contributions of this paper include (1) to propose a practical privacy-preserving protocol for linear multiple regression with vertically partitioned datasets, (2) to show the feasibility of the proposed  ...  As for the data-mining algorithm, we focus on a linear multiple regression that can be used to identify the most significant factors among many possible variables, such as the history of many diseases.  ...  Fig. 8 8 Processing Time for Horizontal Partition Linear Regression (N = 1, n = 257,997). Fig. 9 9 Processing Time for Vertical Partition Linear Regression (N = 1, n = 10,000, M = 1).  ... 
doi:10.2197/ipsjjip.26.638 fatcat:k5tals2uwrh2taemutigfmj7qa

Secret Sharing based Secure Regressions with Applications [article]

Chaochao Chen, Liang Li, Wenjing Fang, Jun Zhou, Li Wang, Lei Wang, Shuang Yang, Alex Liu, Hao Wang
2020 arXiv   pre-print
To make a balance between technical improvements and security limitations, we implement secure and scalable protocols for multiple data holders to train linear regression and logistic regression models  ...  On the other hand, data security concerns may arise for both data holders and data providers due to commercial or sociological concerns.  ...  ) linear regressions on Blog dataset.  ... 
arXiv:2004.04898v1 fatcat:nf4kr3bpmza2fopzemafqhp52e

Secure computation with horizontally partitioned data using adaptive regression splines

Joyee Ghosh, Jerome P. Reiter, Alan F. Karr
2007 Computational Statistics & Data Analysis  
Secure computation protocols enable the owners to compute parameter estimates for some statistical models, including linear regressions, without sharing individual records' data.  ...  When several data owners possess data on different records but the same variables, known as horizontally partitioned data, the owners can improve statistical inferences by sharing their data with each  ...  Secure regression in the vertically partitioned data setting-when data owners possess different variables on the same subjects-faces similar model prespecification dilemmas.  ... 
doi:10.1016/j.csda.2006.10.013 fatcat:gw7z3o52p5b6rkzy54qpwagycy

Valid Statistical Analysis for Logistic Regression with Multiple Sources [chapter]

Stephen E. Fienberg, Yuval Nardi, Aleksandra B. Slavković
2009 Lecture Notes in Computer Science  
We focus mainly on logistic regression, but the method and tools described may be applied essentially to other statistical models as well.  ...  In the paper, we propose an approach that gives full statistical analysis on the combined database without actually combining it.  ...  This is equally true even for linear regression on pure vertically partitioned data, e.g., see [14] .  ... 
doi:10.1007/978-3-642-10233-2_8 fatcat:5e4spq2lifbozcjbqheopjr5ju

"Secure" Logistic Regression of Horizontally and Vertically Partitioned Distributed Databases

Aleksandra B. Slavkovic, Yuval Nardi, Matthew M. Tibbits
2007 Seventh IEEE International Conference on Data Mining Workshops (ICDMW 2007)  
We describe "secure" Newton-Raphson protocol for binary logistic regression in the case of horizontally and vertically partitioned databases using secure-mulity party computation.  ...  We draw from both PPDM and SDL paradigms, and address the problem of performing a "secure" logistic regression on pooled data collected separately by several parties without directly combining their databases  ...  This is equally true even for linear regression on pure vertically partitioned data, e.g., see [19] .  ... 
doi:10.1109/icdmw.2007.114 dblp:conf/icdm/SlavkovicNT07 fatcat:l2o63o7tkzbxlaztkz3mch5qai

Achieving Differential Privacy in Vertically Partitioned Multiparty Learning [article]

Depeng Xu, Shuhan Yuan, Xintao Wu
2019 arXiv   pre-print
Evaluation on real-world and synthetic datasets for linear and logistic regressions shows the effectiveness of our proposed method.  ...  Our method needs only one round of noise addition and secure aggregation.  ...  Linear Regression For linear regression, we first evaluate our method on two real-world datasets.  ... 
arXiv:1911.04587v1 fatcat:knjswozvvfggvlj2hbhvp6a4o4

Privacy-Preserving Generalized Linear Models using Distributed Block Coordinate Descent [article]

Erik-Jan van Kesteren, Chang Sun, Daniel L. Oberski, Michel Dumontier, Lianne Ippel
2019 arXiv   pre-print
Without leaking information, our method performs as well on vertically partitioned data as existing methods on combined data -- all within mere minutes of computation time.  ...  In this paper, we greatly extend the range of analyses available for vertically partitioned data, i.e., data collected by separate parties with different features on the same subjects.  ...  Acknowledgments: We thank Ayoub Bagheri for his helpful comments on an earlier version of this manuscript.  ... 
arXiv:1911.03183v1 fatcat:zudfmjaxwbcrfkxxdbm2uwt2ry

Distributed Outsourced Privacy-Preserving Gradient Descent Methods among Multiple Parties

Zuowen Tan, Haohan Zhang, Peiyi Hu, Rui Gao
2021 Security and Communication Networks  
However, the dataset or the target function's confidentiality may not be kept in secret during computations. Thus, security threats and privacy risks arise.  ...  To address the data and model's privacy mentioned above, we present two new outsourced privacy-preserving gradient descent (OPPGD) method schemes over horizontally or vertically partitioned data among  ...  when the dataset is vertically partitioned.  ... 
doi:10.1155/2021/8876893 doaj:3acb0f2a817e45ae8c62c0df20d4f06d fatcat:orumjhurmbc67hrv6lezpumd3e

FedV: Privacy-Preserving Federated Learning over Vertically Partitioned Data [article]

Runhua Xu, Nathalie Baracaldo, Yi Zhou, Ali Anwar, James Joshi, Heiko Ludwig
2021 arXiv   pre-print
To close this gap, we propose FedV, a framework for secure gradient computation in vertical settings for several widely used ML models such as linear models, logistic regression, and support vector machines  ...  However, many real scenarios follow a vertically-partitioned FL setup, where a complete feature set is formed only when all the datasets from the parties are combined, and the labels are only available  ...  Each dataset is partitioned vertically and equally according to the numbers of parties in all experiments.  ... 
arXiv:2103.03918v2 fatcat:5gkqne5wxjdxrbnvl7gyvtlrkq

Privacy-Preserving Methods for Vertically Partitioned Incomplete Data

Yi Deng, Xiaoqian Jiang, Qi Long
2021 AMIA Annual Symposium Proceedings  
In this paper, we propose a privacy- preserving distributed analysis framework for handling missing data when data are vertically partitioned.  ...  The proposed framework for handling vertically partitioned incomplete data is substantially more privacy-preserving than methods that require pooling of the data, since no individual-level data are shared  ...  In Stage 2, we establish a weighted distributed linear regression model in this vertically partitioned setting, with the weights obtained from Stage 1.  ... 
pmid:33936407 pmcid:PMC8075536 fatcat:x2m2bcgpmjez7amqjnh274se5a

Privacy-Preserving Methods for Vertically Partitioned Incomplete Data [article]

Yi Deng, Xiaoqian Jiang, Qi Long
2020 arXiv   pre-print
In this paper, we propose a privacy-preserving distributed analysis framework for handling missing data when data are vertically partitioned.  ...  The proposed framework for handling vertically partitioned incomplete data is substantially more privacy-preserving than methods that require pooling of the data, since no individual-level data are shared  ...  In Stage 2, we establish a weighted distributed linear regression model in this vertically partitioned setting, with the weights obtained from Stage 1.  ... 
arXiv:2012.14954v1 fatcat:pwum2gxscfembhebokjj5u3iry

When Homomorphic Encryption Marries Secret Sharing: Secure Large-Scale Sparse Logistic Regression and Applications in Risk Control [article]

Chaochao Chen, Jun Zhou, Li Wang, Xibin Wu, Wenjing Fang, Jin Tan, Lei Wang, Alex X. Liu, Hao Wang, Cheng Hong
2021 arXiv   pre-print
In this paper, we first present CAESAR, which combines HE and SS to build secure large-scale sparse logistic regression model and achieves both efficiency and security.  ...  Logistic Regression (LR) is the most widely used machine learning model in industry for its efficiency, robustness, and interpretability.  ...  Other researches proposed to build linear regression [21] and logistic regression [43] under horizontally partitioned data.  ... 
arXiv:2008.08753v2 fatcat:j34yydtupral5ja5pp4gincvoe

Privacy Leakage of Real-World Vertical Federated Learning [article]

Haiqin Weng, Juntao Zhang, Feng Xue, Tao Wei, Shouling Ji, Zhiyuan Zong
2021 arXiv   pre-print
We also experimentally show that the leaked information is as effective as the raw training data through training an alternative classifier on the leaked information.  ...  Yet, their over pursuit of computing efficiency and fast implementation might diminish the security and privacy guarantees of participant's training data, about which little is known thus far.  ...  In vertical federated learning, this dataset does not exist in one place but is composed of the columns of the datasets held by A and B, giving the vertical partition: X = [X A |X B ].  ... 
arXiv:2011.09290v2 fatcat:izxk2vmlvngyfpmeuqxz3evwma

Privacy-Preserving Distributed Linear Regression on High-Dimensional Data

Adrià Gascón, Phillipp Schoppmann, Borja Balle, Mariana Raykova, Jack Doerner, Samee Zahur, David Evans
2017 Proceedings on Privacy Enhancing Technologies  
We propose privacy-preserving protocols for computing linear regression models, in the setting where the training dataset is vertically distributed among several parties.  ...  Like many machine learning tasks, building a linear regression model involves solving a system of linear equations.  ...  Our main contributions are as follows: -Scalable MPC protocols for linear regression on vertically partitioned datasets (Section 3).  ... 
doi:10.1515/popets-2017-0053 dblp:journals/popets/GasconSB0DZE17 fatcat:hpn4a3ulf5dstojfrvjesrjf6y

FedML: A Research Library and Benchmark for Federated Machine Learning [article]

Chaoyang He, Songze Li, Jinhyun So, Xiao Zeng, Mi Zhang, Hongyi Wang, Xiaoyang Wang, Praneeth Vepakomma, Abhishek Singh, Hang Qiu, Xinghua Zhu, Jianzong Wang (+8 others)
2020 arXiv   pre-print
FedML supports three computing paradigms: on-device training for edge devices, distributed computing, and single-machine simulation.  ...  However, existing FL libraries cannot adequately support diverse algorithmic development; inconsistent dataset and model usage make fair algorithm comparison challenging.  ...  Federated datasets for linear models (convex optimization). The linear model category is used for convex optimization experiments such as the ones in [122] and [123] .  ... 
arXiv:2007.13518v4 fatcat:tyoav4xm3bgqbdy2gctnjfeb5i
« Previous Showing results 1 — 15 out of 5,018 results